The arc-sine square-root tranformation was used to stabalize the variance before conducting ANOVA. I’m still unclear about why the logit transformation produces negative values, but I may be applying it incorrectly.
ANOVA and Tukey’s HSD were used to identify significantly different groups at p < 0.05. Comparisons were made on transformed proportions, but data are plotted on original scale. Points show mean, error bars show bootstrapped 95% C.I., and colored points show values at the frame-level.
No formal statistical analysis here, but these are useful for understanding mortalitiy over the seaon. These values are summarized at the site level. I think this speaks to the importance of both germination and survival. If you germinate low, there’s nowhere to go. If you germinate higher, many will die, but difficult to tell if more died on south aspects than others.
Plotted with variable axes, to highlight change within sites.
Plotted with a fixed axis, to allow comparison among sites.
Soil temperature and moisture from April 15 - November 1st 2018. Lines are daily means, ribbons are daily minimum and maximum values.
Air temperature (‘air_temp’, C), precipitation (‘precip’, mm), solar radiation (‘solar’, wm2), vapor pressure defecit (‘vpd’, kPA) at at 1.5 m from ~ June 15 - October 15, 2018. Lines show daily mean, ribbons show daily minimum and maxium.